A multi-scale weakly supervised framework converts underwater point classifications into coarse masks to train UAV coral segmentation models, then refines them via self-training to reach 86.07% pixel accuracy and 52.23% mIoU without pixel-level labels.
The coralscapes dataset: Semantic scene understanding in coral reefs
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2025 2verdicts
UNVERDICTED 2representative citing papers
Survey identifies three drivers—environmental necessity, citizen-science datasets, and researcher migration from terrestrial vision—as transforming underwater perception and spurring advances in weakly supervised and robust AI methods.
citing papers explorer
-
A drone-based framework for coral habitat mapping via weakly supervised segmentation
A multi-scale weakly supervised framework converts underwater point classifications into coarse masks to train UAV coral segmentation models, then refines them via self-training to reach 86.07% pixel accuracy and 52.23% mIoU without pixel-level labels.
-
AI-Driven Marine Robotics: Emerging Trends in Underwater Perception and Ecosystem Monitoring
Survey identifies three drivers—environmental necessity, citizen-science datasets, and researcher migration from terrestrial vision—as transforming underwater perception and spurring advances in weakly supervised and robust AI methods.